Stepper Motor Noise?

[rk396]

I've looked all over and could not find an answer to this, so i'm hoping that someone can shed some light on what's going on here. If this belongs in a different forum, i apologize for posting it here.

I'm building a 3d-printer. I use linuxcnc and a Gacko 540 to control all the steppers. To control the heated nozzle i've built a controller using an arduino uno, AD595 chip, and the LCD shield from adafruit. The arduino reads in the temp from the tc, and switches on/off (using PID) a fet to supply power to the heating element on the nozzle. This all works quite well until i turn on the stepper motors. I can turn on the 540 with no motors connected, and all is well. As soon as i plug in at least 1 motor, and turn on the 540, the TC reading begins to jump all over the place, thus making control of the nozzle non-existent.

The arduino is powered from a laptop. The heater is powered from an external power supply. The 540 is controlled from a second computer, and the steppers are powered from a separate power supply as well. It certainly feels like its a noise issue from the motors, but with all the power supplies being independent, i'm not really sure how noise could be the culprit.

and advise/suggestions would be greatly appreciated.

thanks

 

[Joe G]

not sure myself. sounds like a ground issue to me?
A fun project though, thats what I'm just starting to design myself

 

[tunedwolf]

CNC controlled machinery is notorious for generating electrical noise with large induced spikes from motors and actuators getting into your low level control wiring. You will want at least a common supply ground and shielding for all your wiring on the machine itself and preferably opto isolation on your control inputs as a minimum. Keep all your motor and power wiring neatly trained and well away from your control input wiring and keep all wiring as short as is practical

 

[rk396]

Thanks for the input. I replaced one motor's cabling to a shielded cable. With just that motor plugged in it made a slight improvement, but not enough to say that was the problem. l get the whole isolation of each component, but I'd think that by running off of completely different power supplies that would be isolation enough. The arduino controllers is only controlling the nozzle heater, and is not connected at all to the CNC portion of the system. Yet it does seem like an interference/ground issue.

I'm a mechanical eng posing as an electrical so i'm sure there is something i'm missing.

Joe-G i'd be more then happy to share anything i've done to this point if your interested. I'm sure there are plenty of area's for improvement from what i've got so far.... well obviously otherwise it would have worked right out of the box

 

[Joe G]

thanks. I've been thinking of using a Arduino Mega 2560 for a start. I feel its cheaper then designing and making the main board and then making my own stepper drivers. Right now I've been designing the "ways" and the size for a starter to "play with"

try this link, some good reading.
http://www.microchip.com/forums/f5.aspx
these two came to my mind...
**broken link removed**
**broken link removed**

 

[Joe G]

but with all the power supplies being independent

are they all plugged into the same power strip/outlet?

 

[rk396]

The PC, and stepper power supplies are on the same strip. I thought that might be an issue which is why i was running the arduino off of a battery powered laptop. I think the issue is the noise coming from the steppers. The closer i move the powered motor to the arduino, the more the tc reading fluctuates. I'll take a look at those links when i get home tonight.

 

[Joe G]

also have a look here https://forums.reprap.org/ lots of info

 

[rk396]

Yeah i frequented that site early on in the design. still no luck. I tried filtering the output signal from the ad595, it helped a little but still not a real solution

 

[rk396]

finally figured it out. I had to add a filter to the incoming t/c signal to tha ad595. i hate it when reading posts that talk about the exact problem i'm having.... and then never post a solution. I used a 1k, and 0.1uf cap on the +/- signal wires, and a 10uf cap across them to filter out the noise. I Haven't done the math yet to see why this worked. Thanks for all the help on this.